SwiftExpat

For me this confirms it is cached. I would not waste time in perfmon. I would go with @Kas Ob. and his number 5. I have never read into any of that material but it sounds like a logical boundary that the OS would not allow you to cross.

Have you used Task Manager to watch the cache counter in the Memory tab? It is a good place to start before perfmon. While that disk might look local to you, it is likely on some storage array which could have some block caching. Consider how much RAM is on that server to cache with and how busy is that server when you are executing your programs. Perfmon with some counters set around physical disk, % idle time, read queue length, reads per second and is a good place to start to see when it is actually hitting the physical disk and if it is waiting on disk. Compare your reads per second against run 1 an run 2. Also grab the cache counters and look to see if any of them spike on a second run, this will give you an OS cache look.

They are not nearly the same functionality, stick with TMS. I think of it like this, PY4D loads python39.dll into Delphi at the api level. At this point you have an execution engine. As a developer you want / need to wrap your Delphi API to make it python friendly / usable. A justification would be to python enable your end users as some sort of marketing selling point. I fear you would be teaching / debugging your end user python which is copy paste from the internet. It is valid large investment if you want to grow out of your application something like this where the Wing IDE is interacting with Blender. Look at this as an example and just imagine the work to get there. https://wingware.com/doc/howtos/blender

I do not have a solution for this, but have some thoughts to explore this further: Do you really need the IDE embedded in your application or could you treat it as a remote execution / debug scenario? Which code, Delphi or the Python script, do you need the user to be able to watch / step into? --how deep does the customer really need to watch into your Delphi code, first public level only? Could you use PY4D to host the interpreter and then remote attach to it with the IDE? --this is probably the path to aim for, it may require more wrapping of your Delphi code, but then you have options on the IDE side.

Another way to solve the problem, when you need to add / modify to path, order is likely important. Just run the command in python: import sys sys.path.insert(0, 'path/to/script') The engine has a property, PythonEngine.InitScript: PythonEngine.InitScript.Add('import sys'); PythonEngine.InitScript.Add('sys.path.insert(0, 'path/to/script')'); The benefit is when you attempt to run / unit test in python and need to modify the path. And if you use PyScripter as your python dev tool the approach will be reusable.

Christian, It sounds a lot like desktop heap limitation, most likely on the terminal server. A summary document to describe the problem is here: https://techcommunity.microsoft.com/t5/ask-the-performance-team/sessions-desktops-and-windows-stations/ba-p/372473 The registry reference is here: https://docs.microsoft.com/en-us/troubleshoot/windows-server/performance/desktop-heap-limitation-out-of-memory the fix is in step 4: In the Value data section of the Edit String dialog box, locate the SharedSection entry, and then increase the second value and the third value for this entry. You can bump those values up 512 at a time and see if it helps. Unfortunately it is a reboot each time. For a more scientific approach, I found this from an MVP. https://thundaxsoftware.blogspot.com/2011/10/monitoring-desktop-heap-memory-and.html

I should clarify, ongoing defects considered and compared, Delphi has no more stability problems on my AMD vs Intel laptop. I acknowledge that if the hardware gives me perceived stability, it is likely masking race conditions. My Intel laptop is not likely be more than 10-15% slower in this workload, both are SSD as David points to. Laptops suffer from heat, as soon as the processor gets hot it starts to throttle down. The new desktop has plenty of cooling so it stays at full throttle. Your goal with an AMD build has to be keep it cool to maintain full core speed. So the speed gain is likely staying at full throttle, with a little extra clock speed. If you are looking at a 5600 you should be running more than just Delphi to actually use those 6 cores and the cache you are buying.

I have been running 10.4.2 for 2 months on a AMD 5600X with no stability problems. Mine is a desktop build, so cooling is easy to keep under control. The IDE is very responsive, LSP definitely performs better compared to my laptop, Compile time is 20-30% improved. IDE stability is likely linked to the graphics card / driver underneath more than the processor. I did go very low end with an NVIDIA GT 1030 from a good supplier ASUS.

I am running 11.2.3 Big Sur on virtual box, XCode 11.5. The downside is there is little graphics acceleration, so do not expect to test a UI application at real speed. I would recommend 8GB ram for the VM and 4 procs. You can experiment with it from here, results will vary with hardware, just remeber to add the extras in VBOX: https://www.soupbowl.io/2020/04/macos-in-virtualbox/ https://www.wikigain.com/create-macos-big-sur-iso-image/

Hello Rolf, There is a file in the directory with the suffix, should be python35._pth that needs to be modified to import site. Here are the references I used to get past the issue: https://dev.to/fpim/setting-up-python-s-windows-embeddable-distribution-properly-1081 https://www.christhoung.com/2018/07/15/embedded-python-windows/